Response to induction chemotherapy is one of the most important predictors of outcome in acute myeloid leukemia (AML) as well as cytogenetics and molecular genetics. Measurement of minimal residual disease (MRD) by flow cytometry is an informative method for assessment of initial treatment response, but the heterogeneity of leukemia-associated antigens and antigen shifts during treatment limit its sensitivity and specificity. We prospectively evaluated the prognostic prevalence of MRD monitoring using multi-color flow cytometry in children with AML treated on the Japanese Pediatric Leukemia/Lymphoma Study Group (JPLSG) AML-05 trial.

Patients and methods

From January 2007 to October 2010, 34 children with newly diagnosed de novo AML were enrolled on the AML-MRD study conducted by the Tokyo Children's Cancer Study Group. The median age at the diagnosis was 8 years (1 month- 15 years), and 17 patients were boys and 17 were girls. They were all treated with the JPLSG AML-05 trial. In AML-05, enrolled patients received two induction courses and those achieving a complete remission (CR) were treated according to risk classification at which all the core binding factor-AML cases and good initial response based on morphology after induction 1 were assigned to low risk (LR) group, cases presented with FMS-like tyrosine kinase 3 internal tandem duplication (FLT3-ITD), unfavorable cytogenetics including monosomy7, 5q-,t(16;21), and with poor initial response were assigned to high risk (HR) group, the others were assigned to intermediate risk (IR) group. Allogeneic hematopoietic stem cell transplantation (HSCT) was indicated for the HR patients after the third or later treatment courses. Among the 34 patients enrolled on the AML-MRD study, 8 presented with LR, 22 with IR, and 4 with HR regarding the cytogenetics and FLT3-ITD status. MRD of 63 bone marrow samples were analyzed after induction 1 (BMA-2) and induction 2 (BMA-3) by four-color flow cytometry using 9 AML-associated antigens. A threshold level for MRD-positivity was set at the point of 0.1%.


Sixty-two (98.4%) of 63 bone marrow samples were evaluable for MRD. Thirteen (39.4%) of 33 samples and 8 (27.6%) of 29 showed MRD-positivity at BMA-2 and BMA-3 respectively. Among the patients with MRD-positivity, 12 at BMA-2 and 7 at BMA-3 were diagnosed as achieving CR by morphology. MRD was associated neither specific FAB subtype nor white blood cell count at diagnosis, but all 3 patients with FLT3-ITD showed the MRD-positivity at BMA-2. Although 3-year probability of event free survival (3-yr EFS) at BMA-2 or BMA-3 was similar between patients with and without MRD; 53.8% (n= 13) vs 70.0% (n= 20) (p=0.30) and 50.0% (n= 8)vs 62.0%(n= 21) (p=0.36), respectively, 3-yr EFS of those with MRD at BMA-2 or BMA-3 was significantly worse compared to those without MRD; 33.3% (n= 9) vs 83.3% (n= 12) (p=0.02) and 20.0%(n= 5) vs 76.9% (n= 13) (p=0.04), respectively, in the group of the IR cytogenetics and negative FLT3-ITD. Multivariate analysis indicated that the MRD detected by multi-color flow cytometry was solely associated with worse outcome in this group.


Highly sensitive detection of MRD by multi-color flow cytometry was possible in children with AML. The present study suggests that MRD monitoring may have a prognostic relevance in childhood AML with the IR cytogenetics and negative FLT3-ITD.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.